University of Zurich, Institute of Pharmacology and Toxicology
Sleep homeostasis denotes a basic principle of sleep regulation. A sleep deficit elicits a compensatory increase in the intensity and duration of sleep, while excessive sleep reduces sleep propensity. Slow waves in the electroencephalogram (EEG), a correlate of sleep intensity, serve as an indicator of sleep homeostasis in nonREM sleep, also referred to as slow-wave sleep in animals.
Homeostasis refers to regulatory mechanisms that maintain the constancy of the physiology of organisms.
The term can be applied to sleep: Sleep has a regulatory system enabling organisms to compensate for the loss of sleep or surplus sleep.
The daily sleep-wake cycle, typical for humans, and the polyphasic sleep-wake cycles in animals are regulated by
- a homeostatic mechanism and
- the circadian system
The two main regulated variables are
- sleep intensity and
- to a lesser extent, sleep duration (or the amount of sleep)
The homeostatic mechanisms regulate sleep intensity, while the circadian clock regulates the timing of sleep.
The intensity component of sleep is slow-wave activity (its level correlates negatively with the threshold to arouse subjects or animals). Slow-wave activity is defined as spectral power of the electroencephalogram (EEG) in the frequency range of approximately 0.5 – 4.0 or 4.5 Hz.
Brief history of the concept
The general concept of homeostasis dates back to two pioneers:
- Claude Bernard (1813-1978) a professor of Physiology at the College of France and the Sorbonne in Paris, and
- Walter Cannon (1871-1945) (Cannon, 1939), Professor at Harvard University.
Both men engaged in physiology, Bernard defining a milieu interieur and Cannon, coining the term homeostasis.
Alexander Borbély, a Swiss pharmacologist and sleep researcher who began his research by studying sleep regulation in laboratory rats under diverse controlled conditions (Borbély & Neuhaus, 1979) applied the concept of homeostasis to sleep regulation.
Sleep homeostasis: regulated balance between sleep and waking. Homeostatic mechanisms counteract deviations from an average reference level of sleep (Borbély, 1980) .
Based on his seminal experiments, Borbély postulated a two-process model of sleep homeostasis (Borbély 1980), and extended his concept to human sleep (Borbély 1982). Later, Serge Daan, Domien Beersma from the University of Groningen (The Netherlands) and Alexander Borbély, formulated a quantitative model (Daan et al, 1984). This collaborative work led to the formulation of a simple model, the two-process model of sleep regulation which inspired many researches to test its tenets as well as its applicability to human sleep disorders:
A Process S, the homeostatic process, increases as an exponential saturating function during waking and decreases as an exponential function during sleep. Slow-wave activity in nonREM sleep (sleep is subdivided into rapid-eye movement sleep, REM sleep and nonREM sleep) is the marker for the decrease of Process S.
Seminal experimental data demonstrating the existence of sleep homeostasis
Early experiments had shown that sleep deprivation has major effects on the homeostatic regulation of sleep, but has only minor effects or no effect on the circadian pacemaker. Sleep deprivation invariably leads to an increase in slow-wave activity during recovery sleep. The potential confound that the sleep deprivation induced activation can be stressful, and therefore unspecific factors not directly related to the loss of sleep could be responsible for the increase of slow-waves during recovery sleep was excluded in many studies:
For example, early experiments in two different rat strains (Alan Rechtschaffen, Chicago and Alexander Borbély, Zurich in 1979) showed that doubling the rotation rate of a slowly turning cylinder used to sleep deprive the animals had no additional effect on recovery, and that rats which had no circadian organization of their sleep-wake cycle still showed a compensatory increase of slow waves during recovery from the sleep deprivation (Tobler, Gross & Borbély, 1983).
The evolutionary advantage of developing an intensity dimension of sleep, provided sleep with a relative independence from the circadian system allowing organisms a more flexible adaptation to changes in sleep, than the strictly controlled timing of sleep within the time constraints set by the circadian pacemaker.
Elegant experiments in human subjects
1. Healthy young men slept at different times of day while their sleep was being recorded. A higher level of EEG slow-waves occurred the later in the day they took their nap or in other words, the longer they had been awake since ending last night’s sleep (Dijk, Beersma, Daan, 1987).
2. Similarly, persons taking a 2-h nap in the evening, simulating what often occurs during a normal day when people fall asleep when relaxing after coming home from work, had a lower level of slow-waves during the subsequent nights sleep (Werth et al, 1996).
These studies demonstrated a predictable increase of sleep pressure as a function of the duration of the previous waking interval.
The polyphasic sleep-wake cycle of animals is an ideal feature of animal sleep to examine whether keeping animals awake, in this case many different mouse strains, hamsters, rats, squirrels and even cats, for a varying amount of hours, leads to a predictable change in sleep intensity (reviewed in Tobler, 2005). It did, and the results obtained in the different species were consistent: Slow-wave activity increases as a function of the duration of prior wakefulness.
Later studies extended the variables reflecting sleep homeostasis
- Very short awakenings from sleep (brief awakenings), typical for most animals, decrease when sleep intensity is high (Franken et al, 1991).
- Sleep fragmentation decreases in humans as sleep intensity increases.
Elaboration and applications of the two-process model
The original model was able to account for such diverse phenomena as:
- recovery from sleep deprivation
- circadian phase dependence of sleep duration
- sleep during shift work
- sleep fragmentation during continuous bed rest, and
- internal desynchronization in the absence of time cues.
The two-process model triggered numerous experimental studies to test its predictions and was used to predict the response of habitual short and long sleepers to sleep deprivation.
In a later version of the model it is the change of Process S, and not its level, which is proportional to the momentary amount of SWA. The elaborated model addressed not only the global changes of SWA as represented by Process S, but also the changes within nonREM sleep episodes (Achermann et al., 1993).
This elaborated model successfully predicted the magnitude of the intra-night rebound after selective SWS deprivation (nonREM sleep stages 3 and 4 in humans) in the first 3 h of sleep. Also the occurrence of late SWA peaks during extended sleep could be simulated. The simulations demonstrated that the elaborated model accounts in quantitative terms for empirical data and predicts the changes induced by the prolongation of waking or sleep.
This model was also used to simulate the dynamics of SWA in an experimental protocol with an
- early evening nap and
- the effect of changes in REM sleep latency on the time course of slow-wave activity.
Finally, not only the timing of sleep but also the changes in daytime vigilance are governed by the interaction of Process S and C. The rising homeostatic sleep pressure during waking seems to be compensated by the declining circadian sleep propensity. Conversely, during sleep the rising circadian sleep propensity may serve to counteract the declining homeostatic sleep pressure, thereby ensuring the maintenance of sleep.
Sleep in clinically depressed patients exhibits pathognomonic changes (prolonged sleep latency, a shallow fragmented sleep process, precocious awakening in the morning). Slow wave sleep is typically reduced. Sleep deprivation for one night exerts an immediate antidepressant effect that is short lived. It was hypothesized that sleep regulation (Process S) is deficient in depression (Borbély and Wirz-Justice, 1982). The antidepressant effect of sleep deprivation was attributed to the increased level of Process S attained by prolonging wakefulness.
Sleep homeostasis in narcoleptic patients is functional (Khatami et al., in press). However, the decline of Process S appears to be steeper in patients. This may be related to the increased number and longer duration of short wake episodes in the second and third sleep cycle. Thus, the decline of S is not exponential and may be better approximated by the elaborated model.
The increase of homeostatic sleep pressure during wakefulness is faster in prepubertal or early pubertal children compared with mature adolescents, while the decrease of Process S is similar in both developmental groups (Jenni et al., 2005). These age-related differences indicate that the brain reaches its capacity to generate slow waves with less time awake in the young than in the mature brain.
Animals ranging from mammals to birds and even to invertebrates show compensatory mechanisms after sleep loss (Tobler, 2005). The first such demonstration was in cockroaches and later in scorpions (Campbell and Tobler, 1984; Tobler, 2005). However, the strict relationship between the amount of sleep lost and the degree of compensation has been demonstrated for a few mammals, such as rats, mice, hamsters, squirrels and humans and the fruit fly.
Especially the discovery that also insects do compensate for the loss of sleep, inspired research on sleep regulation in thousands of mutants of the fruit fly Drosophila, where the genetic mechanisms of the compensatory process can be investigated (cross reference).
Neuronal mechanisms underlying sleep homeostasis
Sleep deprivation causes behavioral, physiological and molecular changes. Despite considerable knowledge about the neuronal mechanisms enabling the transition from wakefulness to sleep (cross reference), and the synchronization of EEG waves in the cortex, the mechanisms leading to the intensity increase are still not fully understood.
A common belief is the existence of a sleep factor (or perhaps several sleep factors) accumulating during waking and dissipating during sleep. Many neurotransmitters and neuropeptides must be involved in sleep regulation, but one such substance, adenosine, a neurotransmitter, is more and more at the center of attention. Manipulating the adenosine system leads to changes in sleep (reviewed in Basheer et al, 2004). Especially an adenosine antagonist, caffeine, is the world-wide most popular wakefulness inducing and maintaining substance. Caffeine reduces slow-wave activity in the subsequent sleep episode and caffeine consumption during prolonged wakefulness counteracts the typical effects of sleep deprivation on the waking and sleep EEG (Landolt et al, 2004).
Open questions and perspectives
It is still unresolved whether REM sleep has a homeostatic regulatory component of its own. REM sleep loss does lead to an increase in the tendency to enter REM sleep, and its loss is compensated up to a certain extent only, with some species differences. However, in contrast to nonREM sleep which has an intensity dimension,
there is no evidence for an intensity dimension of REM sleep.
Recent experiments used exquisite, selective manipulations activating specific brain regions during sleep. In rats and mice, cutting whiskers on one side of the snout and encouraging spontaneous stimulation of the remaining whiskers by placing the animals in an enriched environment, led to a selective increase of slow waves over the stimulated brain region during sleep (Vyazovskiy et al, 2004).
An early study in humans showed a similar selective increase in slow-waves over the “stimulated hemisphere” during sleep, after one hand had been vibrated previously for several hours (Kattler et al, 1994).
Recordings making use of EEG topography, placing 256 electrodes on the heads of subjects and subjecting them to a motor learning task before sleep, led to a selective increase of slow-waves over the particular brain region where neurons had been stimulated during the learning task (Huber et al., 2004). Furthermore, immobilization of the arm led to a decrease of slow-wave activity in the corresponding motor areas (Huber et al., 2006). The studies in the rat were developed further, showing a correlation between increases in the brain derived nerve growth factor (BDNF) during sleep and spontaneous exploratory waking activities.
These studies pursue the hypothesis that sleep may be rescaling neuronal connections and synapses according to their previous use (Tononi and Cirelli, 2006).
Cannon WB: The Wisdom of the Body, New York, WW Norton, 1939
Basheer R., Strecker RE., Thakkar MM. and McCarley RW. Adenosine and sleep-wake regulation. Progress in Neurobiology 73 (6) 2004.
Borbély AA. and Neuhaus HU. Sleep deprivation: Effects on sleep and EEG in the rat. J. comp Physiol A 133 (1979) 71 87.
Borbély AA. Sleep: circadian rhythm versus recovery process. In: Koukkou M., Lehmann D. and Angst J. (Eds.) Functional states of the brain: their determinants. Elsevier, Amsterdam, 1980, p. 151 161.
Borbély AA., Baumann F., Brandeis D., Strauch I. and Lehmann D. Sleep deprivation; effect on sleep stages and EEG power density in man. Electroenceph clin. Neurophysiol 51 (1981) 483 493.
Borbély AA. A two process model of sleep. Human Neurobiol 1 (1982) 195 204.
Borbély AA. and Wirz-Justice A. Sleep, sleep deprivation and depression. A hypothesis derived from a model of sleep regulation. Human Neurobiol 1 (1982) 205 210.
Borbély AA. and Achermann P. Sleep homeostasis and models of sleep regulation. In: Principles and Practice of Sleep Medicine, Kryger M.H., Roth T., Dement W.C. (Eds.), Elsevier Saunders, Philadelphia (2005) p. 405-417.
Campbell SS. and Tobler I. Animal sleep: A review of sleep duration across phylogeny. Neurosci Biobehav Rev 8 (1984) 269-300.
Daan S., Beersma DGM. and Borbély AA. Timing of human sleep: recovery process gated by a circadian pacemaker. Am J Physiol 246 (1984) R161 R183.
Franken P., Dijk DJ., Tobler I. and Borbély AA. Sleep deprivation rats: effects on EEG power spectra, vigilance states, and cortical temperature. Am J Physiol 261 (1991) R198-R208.
Friedmann L., Bergmann BM and Rechtschaffen A. Effcets of sleep deprivation on sleepiness, sleep intensity, and subsequent sleep in the rat. Sleep 1 (1979) 369-391.
Dijk DJ., Beersma DG and Daan S. EEG power density during nap sleep: reflection of an hourglass measuring the duration of prior wakefulness. J Biol Rhythms 2 (1987) 207-209
Huber R. Ghilardi MF., Massimini M. and Tononi G. Local sleep and learning Nature 430 (2004) 78-81.
Huber R., Ghilardi MF, Massimini M., Ferrarelli F., Riedner BA., Peterson MJ. and Tononi G. Arm immobilization causes cortical plastic changes and locally decreases sleep slow wave activity. Nature Neurosci 9 (2006) 1169-1176.
Jenni OG., Achermann P. and Carskadon MA. Homeostatic sleep regulation in adolescents. SLEEP 28 (2005) 1446-1454.
Kattler H., Dijk DJ., Borbély AA. Effect of unilateral somatosensory stimulation prior to sleep on the sleep EEG in humans. J Sleep Res 3 (1994) 159-164.
Khatami R., Landolt H.-P., Achermann P., Rétey J.V, Werth E., Mathis J., Bassetti, CL. Insufficient non-REM sleep intensity in narcolepsy-cataplexy. SLEEP (2007) in press.
Landolt HP., Rétey JV., Tönz K., Gottselig JM. Khatami R., Buckelmüller I. and Achermann P. Caffeine attenuates waking and sleep electroencephalographic markers of sleep homeostasis in humans. Neuropsychopharmacology 29, (2004) 1933-1939.
Tobler I., Borbély AA. and Groos G. The effect of sleep deprivation on sleep in rats with suprachiasmatic lesions. Neurosci Lett 42 (1983) 49 54.
Tobler I. Phylogeny of sleep regulation. Kryger MH., Roth T. III, Dement, WC. (Eds.) Principles and practice of sleep medicine. Elsevier/Saunders, Philadelphia, (2005), p. 77-90.
Tononi G. and Cirelli C. Sleep and synaptic homeostasis: a hypothesis. Brain Res Bull 62 (2003) 143-150.
Tononi G. and Cirelli C. Sleep function and synaptic homeostasis. Sleep Medicine Rev 10 (2006) 49-62.
Vyazovskiy V.V., Welker E., Fritschy J.M., Tobler I. Regional pattern of metabolic activation is reflected in the sleep EEG after sleep deprivation combined with unilateral whisker stimulation in mice. Eur J Neurosci 20 (2004) 1363-1370.
Werth E., Achermann P., Borbély A.A. Brain topography of the human sleep EEG: antero-posterior shifts of spectral power. NeuroReport 8 (1996) 123-127.